There's a long tail for hammers.

The massive burst in innovation that we've seen in the past several decades is caused by a bunch of people making a bunch of new hammers for a bunch of different problem types. In addition, there are certain hammers -- like statistical methods or X-rays or computer networking -- that can be applied to a ranging scope of problems that no one ever predicted before.

Seth is right, you have to find the right tool for the job. But your job is also to imagine and construct a new hammer.